HESPKRini: KUOAMUS PROTEUS. 1391 



Geographical distribution (27 : l). This is a tropical species, being 

 f'oiiiui ill tin' g^rcuti'st :il)iMul;ince tlirouglioiit every part of tiio nortiieni 

 tropics as well as in Suutii Aiiieriea soutli of tiie equator, probably tbrougli- 

 out the entire extent of tiie tropics tliere, possibly oven further, but this I 

 have no means of verifying from specimens. It enters our country at its 

 extreme southern limits. I do not find any report of its occurrence far 

 inland ; neither Gosse nor Grote mention it fnnii central Alabama, but it 

 is found from southern Te.xas where it is moderately common (Aaron) and 

 along the Texan l)oundary (Captain Pope) to Florida, where it extends 

 from the Keys (Maynai'd, Palmer) to Apalachicola (Chapman) and 

 Jacksonville (Palmer.) All the Mexican localities from wiiich I have 

 seen it or noted it were mostly outside the limits of our map and I have no 

 proof of its occurrence on the west coast within our own territory, though 

 its distribution in the east would seem to render it probable that it would 

 be found there ; for along the Atlantic sea-board it extends a considerable 

 way to the north, being found in Wilmington Island (Oemler) and the 

 interior of Georgia (Abbot) and in tolerable abundance in South Carolina 

 (Scudder). It has occured but once in Maryland, to Mr. Uhler's knowl- 

 edge, and recently a number of specimens have been taken in Central Park, 

 New York by Mr. Elliot. 



A few specimens have also been taken about New Haven, Conn. 

 (Smith), tlic only locality where it has been observed in New England. 



Oviposition. According to Comstock, "the eggs are laid in small 

 clusters of from four to six each." Wittfeld writes that they are "de- 

 posited on upper and under side of many Leguminosae, mostly on the under 

 side and mostly in clusters of from two to twenty." They are not, how- 

 ever, "with no discernible markings," as Comstock states, but vertically 

 though rather feebly ribbed. 



Food plants. This caterpillar feeds upon Leguminosae of different 

 tribes ; such as Wistaria frutescens D. C. and Desmodium viridiflorum 

 Beck (Chapman), Phaseolus perennis Walt., "kidney bean" (Abbot, 

 Chapman), Clitoria mariana Linn., the butterfly pea or "wild pea vine" of 

 Abbot (Abbot and Smith), and in Cuba on C. ternatea and other 

 species of the genus (Gundlach). According to Comstock, however, it 

 does not confine itself to leguminous plants but feeds also on Cruciferae, 

 doing in Florida much damage to the crops, particularly beans, turnips 

 and cabbage ! 



Habits of the larva. Like all the caterpillars of the Hesperidi yet 

 known, those of this insect construct nests to live in. " Their method of 

 work, "says Comstock, "was for each to cut a slit into the leaf from the 

 edge, and roll the Hap thus formed around its body, working from the 

 inside of this roll, with its soft parts perfectly protected. In the garden 

 of the Brock House, Enterprise, almost every plant was badly rao-ged in 



